Avoid -ve color values in text-span dicts and add alpha value.

julian-smith-artifex-com · julian-smith-artifex-com · commit bbae37edd2b1 · 2025-01-20T17:56:10.000Z
Color value from MuPDF now contains alpha value in top 8 bits so appeared -ve if alpha sets top bit. Also added alpha value to the dict. Addresses #4139.
diff --git a/docs/textpage.rst b/docs/textpage.rst
@@ -292,15 +292,22 @@ descender       descender of the font *(float)*
 size            font size *(float)*
 flags           font characteristics *(int)*
 char_flags      char characteristics *(int)*
-color           text color in sRGB format *(int)*
+color           text color in sRGB format 0xRRGGBB *(int)*.
+alpha           text opacity 0..255 *(int)*.
 text            (only for :meth:`extractDICT`) text *(str)*
 chars           (only for :meth:`extractRAWDICT`) *list* of character dictionaries
 =============== =====================================================================
 
+|history_begin|
+
+*(New in version 1.25.3.0):* Added *"alpha"* item.
+
 *(New in version 1.16.0):* *"color"* is the text color encoded in sRGB (int) format, e.g. 0xFF0000 for red. There are functions for converting this integer back to formats (r, g, b) (PDF with float values from 0 to 1) :meth:`sRGB_to_pdf`, or (R, G, B), :meth:`sRGB_to_rgb` (with integer values from 0 to 255).
 
 *(New in v1.18.5):* *"ascender"* and *"descender"* are font properties, provided relative to :data:`fontsize` 1. Note that descender is a negative value. The following picture shows the relationship to other values and properties.
 
+|history_end|
+
 .. image:: images/img-asc-desc.*
    :scale: 60
 
diff --git a/src/__init__.py b/src/__init__.py
@@ -16355,7 +16355,7 @@ def __init__(self, rhs=None):
                 if mupdf_version_tuple >= (1, 25, 2):
                     self.char_flags = rhs.char_flags
                 self.font = rhs.font
-                self.color = rhs.color
+                self.argb = rhs.argb
                 self.asc = rhs.asc
                 self.desc = rhs.desc
             else:
@@ -16364,7 +16364,7 @@ def __init__(self, rhs=None):
                 if mupdf_version_tuple >= (1, 25, 2):
                     self.char_flags = -1
                 self.font = ''
-                self.color = -1
+                self.argb = -1
                 self.asc = 0
                 self.desc = 0
         def __str__(self):
@@ -16402,9 +16402,9 @@ def __str__(self):
             style.char_flags = ch.m_internal.flags
         style.font = JM_font_name(mupdf.FzFont(mupdf.ll_fz_keep_font(ch.m_internal.font)))
         if mupdf_version_tuple >= (1, 25):
-            style.color = ch.m_internal.argb
+            style.argb = ch.m_internal.argb
         else:
-            style.color = ch.m_internal.color
+            style.argb = ch.m_internal.color
         style.asc = JM_font_ascender(mupdf.FzFont(mupdf.ll_fz_keep_font(ch.m_internal.font)))
         style.desc = JM_font_descender(mupdf.FzFont(mupdf.ll_fz_keep_font(ch.m_internal.font)))
 
@@ -16414,7 +16414,7 @@ def __str__(self):
                     and (style.char_flags & ~mupdf.FZ_STEXT_SYNTHETIC)
                         != (old_style.char_flags & ~mupdf.FZ_STEXT_SYNTHETIC)
                     )
-                or style.color != old_style.color
+                or style.argb != old_style.argb
                 or style.font != old_style.font
                 ):
             if old_style.size >= 0:
@@ -16446,7 +16446,9 @@ def __str__(self):
             if mupdf_version_tuple >= (1, 25, 2):
                 span[dictkey_char_flags] = style.char_flags
             span[dictkey_font] = JM_EscapeStrFromStr(style.font)
-            span[dictkey_color] = style.color
+            span[dictkey_color] = style.argb & 0xffffff
+            if mupdf_version_tuple >= (1, 25, 0):
+                span['alpha'] = style.argb >> 24
             span["ascender"] = asc
             span["descender"] = desc
 
diff --git a/src/extra.i b/src/extra.i
@@ -1920,7 +1920,6 @@ static void jm_trace_text_span(
             << " fsize=" << fsize
             << " linewidth=" << linewidth
             << "\n";
-    
     dict_setitem_drop(span_dict, dictkey_color, Py_BuildValue("fff", rgb[0], rgb[1], rgb[2]));
     dict_setitem_drop(span_dict, dictkey_size, PyFloat_FromDouble(fsize));
     dict_setitemstr_drop(span_dict, "opacity", PyFloat_FromDouble((double) alpha));
@@ -3040,7 +3039,7 @@ mupdf::FzRect JM_make_spanlist(
     struct char_style
     {
         float size = -1;
-        int flags = -1;
+        unsigned flags = 0;
         
         #if MUPDF_VERSION_GE(1, 25, 2)
         /* From mupdf:include/mupdf/fitz/structured-text.h:fz_stext_char::flags, which
@@ -3052,11 +3051,11 @@ mupdf::FzRect JM_make_spanlist(
         FZ_STEXT_STROKED = 32,
         FZ_STEXT_CLIPPED = 64
         */
-        int char_flags;
+        unsigned char_flags = 0;
         #endif
         
         const char *font = "";
-        unsigned int color = -1;
+        unsigned argb = 0;
         float asc = 0;
         float desc = 0;
     };
@@ -3086,9 +3085,9 @@ mupdf::FzRect JM_make_spanlist(
         #endif
         style.font = JM_font_name(ch.m_internal->font);
         #if MUPDF_VERSION_GE(1, 25, 0)
-            style.color = ch.m_internal->argb;
+            style.argb = ch.m_internal->argb;
         #else
-            style.color = ch.m_internal->color;
+            style.argb = ch.m_internal->color;
         #endif
         style.asc = JM_font_ascender(ch.m_internal->font);
         style.desc = JM_font_descender(ch.m_internal->font);
@@ -3099,7 +3098,7 @@ mupdf::FzRect JM_make_spanlist(
                 #if MUPDF_VERSION_GE(1, 25, 2)
                 || (style.char_flags & ~FZ_STEXT_SYNTHETIC) != (old_style.char_flags & ~FZ_STEXT_SYNTHETIC)
                 #endif
-                || style.color != old_style.color
+                || style.argb != old_style.argb
                 || strcmp(style.font, old_style.font) != 0
                 )
         {
@@ -3135,12 +3134,15 @@ mupdf::FzRect JM_make_spanlist(
             }
 
             DICT_SETITEM_DROP(span, dictkey_size, Py_BuildValue("f", style.size));
-            DICT_SETITEM_DROP(span, dictkey_flags, Py_BuildValue("i", style.flags));
+            DICT_SETITEM_DROP(span, dictkey_flags, Py_BuildValue("I", style.flags));
             #if MUPDF_VERSION_GE(1, 25, 2)
-            DICT_SETITEM_DROP(span, dictkey_char_flags, Py_BuildValue("i", style.char_flags));
+            DICT_SETITEM_DROP(span, dictkey_char_flags, Py_BuildValue("I", style.char_flags));
             #endif
             DICT_SETITEM_DROP(span, dictkey_font, JM_EscapeStrFromStr(style.font));
-            DICT_SETITEM_DROP(span, dictkey_color, Py_BuildValue("i", style.color));
+            DICT_SETITEM_DROP(span, dictkey_color, Py_BuildValue("I", style.argb & 0xffffff));
+            #if MUPDF_VERSION_GE(1, 25, 0)
+            DICT_SETITEMSTR_DROP(span, "alpha", Py_BuildValue("I", style.argb >> 24));
+            #endif
             DICT_SETITEMSTR_DROP(span, "ascender", Py_BuildValue("f", asc));
             DICT_SETITEMSTR_DROP(span, "descender", Py_BuildValue("f", desc));
 
diff --git a/tests/resources/test_4139.pdf b/tests/resources/test_4139.pdf
diff --git a/tests/test_textextract.py b/tests/test_textextract.py
@@ -422,3 +422,28 @@ def test_4147():
                             else:
                                 print(f'            span: {span["flags"]=:#x}')
                                 assert 'char_flags' not in span
+
+
+def test_4139():
+    path = os.path.normpath(f'{__file__}/../../tests/resources/test_4139.pdf')
+    flags = (0
+            | pymupdf.TEXT_PRESERVE_IMAGES
+            | pymupdf.TEXT_PRESERVE_WHITESPACE
+            | pymupdf.TEXT_CID_FOR_UNKNOWN_UNICODE
+            )
+    with pymupdf.open(path) as document:
+        page = document[0]
+        dicts = page.get_text('dict', flags=flags, sort=True)
+        seen = set()
+        for b_ctr, b in enumerate(dicts['blocks']):
+             for l_ctr, l in enumerate(b.get('lines', [])):
+                for s_ctr, s in enumerate(l['spans']):
+                    color = s.get('color')
+                    if color is not None and color not in seen:
+                        seen.add(color)
+                        print(f"B{b_ctr}.L{l_ctr}.S{s_ctr}: {color=} {hex(color)=} {s=}")
+                        assert color == 0, f'{s=}'
+                        if pymupdf.mupdf_version_tuple >= (1, 25):
+                            assert s['alpha'] == 255
+                        else:
+                            assert not 'alpha' in s
