Fix several small correctness bugs

src/__init__.py

@@ -943,6 +943,7 @@ def file_info(self):
 
         fs = mupdf.pdf_dict_get(annot_obj, PDF_NAME('FS'))
 
+        filename = None
         o = mupdf.pdf_dict_get(fs, PDF_NAME('UF'))
         if o.m_internal:
             filename = mupdf.pdf_to_text_string(o)
@@ -2053,7 +2054,7 @@ def make_subarch(entries, mount, fmt):
                 raise ValueError(f'Not a file or directory: {content!r}')
 
         elif is_binary_data(content):
-            assert isinstance(path, str) and path != '' \
+            assert isinstance(path, str) and path != '', \
                     f'Need name for binary content, but {path=}.'
             self._add_treeitem(content, path)
             return make_subarch([path], None, 'tree')
@@ -9482,6 +9483,8 @@ def on_state(self):
         if bstate is None:
             bstate = dict()
         for k in bstate.keys():
+            if bstate[k] is None:
+                continue
             for v in bstate[k]:
                 if v != "Off":
                     return v
@@ -13140,7 +13143,7 @@ def set_language(self, language=None):
             lang = mupdf.fz_text_language_from_string(language)
             assert hasattr(mupdf, 'fz_string_from_text_language2')
             mupdf.pdf_dict_put_text_string(
-                    pdfpage.obj,
+                    pdfpage.obj(),
                     PDF_NAME('Lang'),
                     mupdf.fz_string_from_text_language2(lang)
                     )
@@ -14179,7 +14182,7 @@ def tobytes(self, output="png", jpg_quality=95):
         if idx is None:
             raise ValueError(f"Image format {output} not in {tuple(valid_formats.keys())}")
         if self.alpha and idx in (2, 6, 7):
-            raise ValueError("'{output}' cannot have alpha")
+            raise ValueError(f"'{output}' cannot have alpha")
         if self.colorspace and self.colorspace.n > 3 and idx in (1, 2, 4):
             raise ValueError(f"unsupported colorspace for '{output}'")
         if idx == 7:
@@ -20996,7 +20999,7 @@ def JM_matrix_from_py(m):
     for i in range(6):
         a[i] = JM_FLOAT_ITEM(m, i)
         if a[i] is None:
-            return mupdf.FzRect()
+            return mupdf.FzMatrix()
     return mupdf.FzMatrix(a[0], a[1], a[2], a[3], a[4], a[5])
 
 

src/_apply_pages.py

@@ -137,11 +137,12 @@ def childfn():
         while 1:
             if verbose:
                 pymupdf.log(f'{os.getpid()=}: calling get().')
-            page_num = queue_down.get()
+            item = queue_down.get()
             if verbose:
-                pymupdf.log(f'{os.getpid()=}: {page_num=}.')
-            if page_num is None:
+                pymupdf.log(f'{os.getpid()=}: {item=}.')
+            if item is None:
                 break
+            index, page_num = item
             try:
                 if not document:
                     if stats:
@@ -172,9 +173,9 @@ def childfn():
                 if verbose: pymupdf.log(f'{os.getpid()=}: exception {e=}')
                 ret = e
             if verbose:
-                pymupdf.log(f'{os.getpid()=}: sending {page_num=} {ret=}')
-                
-            queue_up.put( (page_num, ret) )
+                pymupdf.log(f'{os.getpid()=}: sending {index=} {ret=}')
+
+            queue_up.put( (index, ret) )
 
     error = None
 
@@ -206,24 +207,24 @@ def childfn():
             t = time.time()
         if verbose:
             pymupdf.log(f'Sending page numbers.')
-        for page_num in range(len(pages)):
-            queue_down.put(page_num)
+        for index, page_num in enumerate(pages):
+            queue_down.put((index, page_num))
         if stats:
             _stats_write(t, 'Send page numbers')
 
         # Collect results. We give up if any worker sends an exception instead
         # of text, but this hasn't been tested.
         ret = [None] * len(pages)
         for i in range(len(pages)):
-            page_num, text = queue_up.get()
+            index, text = queue_up.get()
             if verbose:
-                pymupdf.log(f'{page_num=} {len(text)=}')
-            assert ret[page_num] is None
+                pymupdf.log(f'{index=} {len(text)=}')
+            assert ret[index] is None
             if isinstance(text, Exception):
                 if not error:
                     error = text
                 break
-            ret[page_num] = text
+            ret[index] = text
 
         # Close queue. This should cause exception in workers and terminate
         # them, but on macos-arm64 this does not seem to happen, so we also

src/table.py

@@ -196,12 +196,15 @@ def rect_in_rect(inner, outer):
 
 def chars_in_rect(CHARS, rect):
     """Check whether any of the chars in CHAR are inside rectangle 'rect'."""
+    # NB: 'rect' (e.g. from page.get_drawings()) and c["x0"]/c["x1"] are in
+    # top-down page space; c["top"]/c["bottom"] are the top-down counterparts
+    # of c["y0"]/c["y1"], which are in PDF-native (bottom-up) space instead.
     return any(
         1
         and rect[0] <= c["x0"]
         and c["x1"] <= rect[2]
-        and rect[1] <= c["y0"]
-        and rect[3] >= c["y1"]
+        and rect[1] <= c["top"]
+        and rect[3] >= c["bottom"]
         for c in CHARS
     )
 

src/utils.py

@@ -210,7 +210,7 @@ def line_text(clip, line):
             lrect |= r  # update line bbox
             # convert distance to previous word to multiple spaces
             dist = max(
-                int(round((r.x0 - x1) / r.width * len(t))),
+                int(round((r.x0 - x1) / r.width * len(t))) if r.width else 0,
                 0 if (x1 == clip.x0 or r.x0 <= x1) else 1,
             )  # number of space characters
 
@@ -498,8 +498,6 @@ def get_text(
     }
     option = option.lower()
     assert option in formats
-    if option not in formats:
-        option = "text"
     if flags is None:
         flags = formats[option]
 
@@ -962,7 +960,10 @@ def get_label_pno(pgNo, labels):
     """
     # Jorj McKie, 2021-01-06
 
-    item = [x for x in labels if x[0] <= pgNo][-1]
+    candidates = [x for x in labels if x[0] <= pgNo]
+    if not candidates:
+        return ""
+    item = candidates[-1]
     rule = rule_dict(item)
     prefix = rule.get("prefix", "")
     style = rule.get("style", "")