summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
Diffstat
-rw-r--r--Jamfile4
-rw-r--r--include/mupdf/rsrc.h15
-rw-r--r--mupdf/font.c58
-rw-r--r--mupdf/fontagl.c (renamed from mupdf/fontagl.h)41
-rw-r--r--mupdf/fontenc.c (renamed from mupdf/fontenc.h)34
-rw-r--r--mupdf/interpret.c2
-rw-r--r--mupdf/type3.c48
-rw-r--r--object/simple.c4
-rw-r--r--test/pdfrip.c4
9 files changed, 114 insertions, 96 deletions
diff --git a/Jamfile b/Jamfile
index f39fda5a..d3240e76 100644
--- a/Jamfile
+++ b/Jamfile
@@ -110,8 +110,10 @@ Library libmupdf :
mupdf/function.c
mupdf/cmap.c
- mupdf/font.c
+ mupdf/fontagl.c
+ mupdf/fontenc.c
mupdf/fontfile.c
+ mupdf/font.c
mupdf/type3.c
mupdf/colorspace.c
mupdf/image.c
diff --git a/include/mupdf/rsrc.h b/include/mupdf/rsrc.h
index 30a059b1..bb29dc8a 100644
--- a/include/mupdf/rsrc.h
+++ b/include/mupdf/rsrc.h
@@ -103,6 +103,21 @@ fz_error *pdf_loadimage(pdf_image **imgp, pdf_xref *xref, fz_obj *obj, fz_obj *s
* CMap and Font
*/
+struct pdf_aglpair { char *name; unsigned short code; };
+extern struct pdf_aglpair pdf_adobeglyphlist[];
+extern int pdf_adobeglyphlen;
+
+void pdf_loadencoding(char **estrings, char *encoding);
+int pdf_lookupagl(char *name);
+
+extern char *pdf_macroman[256];
+extern char *pdf_macexpert[256];
+extern char *pdf_winansi[256];
+extern char *pdf_standard[256];
+extern char *pdf_expert[256];
+extern char *pdf_symbol[256];
+extern char *pdf_zapfdingbats[256];
+
typedef struct pdf_font_s pdf_font;
struct pdf_font_s
diff --git a/mupdf/font.c b/mupdf/font.c
index 231cfc1f..d9683596 100644
--- a/mupdf/font.c
+++ b/mupdf/font.c
@@ -10,9 +10,6 @@
#include FT_FREETYPE_H
#include <freetype/internal/ftobjs.h>
-#include "fontenc.h"
-#include "fontagl.h"
-
static char *basefontnames[14][7] =
{
{ "Courier", "CourierNew", "CourierNewPSMT", 0 },
@@ -163,54 +160,11 @@ static char *cleanfontname(char *fontname)
return fontname;
}
-static void loadencoding(char **estrings, char *encoding)
-{
- char **bstrings = nil;
- int i;
-
- if (!strcmp(encoding, "MacRomanEncoding"))
- bstrings = macroman;
- if (!strcmp(encoding, "MacExpertEncoding"))
- bstrings = macexpert;
- if (!strcmp(encoding, "WinAnsiEncoding"))
- bstrings = winansi;
-
- if (bstrings)
- for (i = 0; i < 256; i++)
- estrings[i] = bstrings[i];
-}
-
-static int aglcode(char *name)
-{
- int l = 0;
- int r = adobeglyphlen;
-
- while (l <= r)
- {
- int m = (l + r) >> 1;
- int c = strcmp(name, adobeglyphlist[m].name);
- if (c < 0)
- r = m - 1;
- else if (c > 0)
- l = m + 1;
- else
- return adobeglyphlist[m].code;
- }
-
- if (strstr(name, "uni") == name)
- return strtol(name + 3, 0, 16);
-
- if (strstr(name, "u") == name)
- return strtol(name + 1, 0, 16);
-
- return -1;
-}
-
static int mrecode(char *name)
{
int i;
for (i = 0; i < 256; i++)
- if (macroman[i] && !strcmp(name, macroman[i]))
+ if (pdf_macroman[i] && !strcmp(name, pdf_macroman[i]))
return i;
return -1;
}
@@ -376,7 +330,7 @@ printf("loading simple font %s\n", basefont);
for (i = 0; i < 256; i++)
{
- estrings[i] = _notdef;
+ estrings[i] = nil;
etable[i] = 0;
}
@@ -388,7 +342,7 @@ printf("loading simple font %s\n", basefont);
goto cleanup;
if (fz_isname(encoding))
- loadencoding(estrings, fz_toname(encoding));
+ pdf_loadencoding(estrings, fz_toname(encoding));
if (fz_isdict(encoding))
{
@@ -396,7 +350,7 @@ printf("loading simple font %s\n", basefont);
base = fz_dictgets(encoding, "BaseEncoding");
if (fz_isname(base))
- loadencoding(estrings, fz_toname(base));
+ pdf_loadencoding(estrings, fz_toname(base));
diff = fz_dictgets(encoding, "Differences");
if (fz_isarray(diff))
@@ -434,7 +388,7 @@ printf(" winansi cmap\n");
for (i = 0; i < 256; i++)
if (estrings[i])
{
- k = aglcode(estrings[i]);
+ k = pdf_lookupagl(estrings[i]);
if (k == -1)
etable[i] = FT_Get_Name_Index(face, estrings[i]);
else
@@ -497,7 +451,7 @@ printf(" builtin encoding\n");
for (i = 0; i < 256; i++)
if (estrings[i])
- utable[i] = aglcode(estrings[i]);
+ utable[i] = pdf_lookupagl(estrings[i]);
else
utable[i] = i;
diff --git a/mupdf/fontagl.h b/mupdf/fontagl.c
index 847c0a21..2b083313 100644
--- a/mupdf/fontagl.h
+++ b/mupdf/fontagl.c
@@ -1,19 +1,44 @@
-/* Name: Adobe Glyph List
+/* Name: Adobe Glyph List
# Table version: 2.0
- # Date: September 20, 2002
+ # Date: September 20, 2002
#
# See http:partners.adobe.com/asn/developer/typeforum/unicodegn.html
#
# Format: Semicolon-delimited fields:
- # (1) glyph name
- # (2) Unicode scalar value
+ # (1) glyph name
+ # (2) Unicode scalar value
*/
-struct aglpair { char *name; int code; };
+#include <fitz.h>
+#include <mupdf.h>
-#define adobeglyphlen (sizeof(adobeglyphlist) / sizeof(struct aglpair))
+int pdf_lookupagl(char *name)
+{
+ int l = 0;
+ int r = pdf_adobeglyphlen;
+
+ while (l <= r)
+ {
+ int m = (l + r) >> 1;
+ int c = strcmp(name, pdf_adobeglyphlist[m].name);
+ if (c < 0)
+ r = m - 1;
+ else if (c > 0)
+ l = m + 1;
+ else
+ return pdf_adobeglyphlist[m].code;
+ }
+
+ if (strstr(name, "uni") == name)
+ return strtol(name + 3, 0, 16);
-static struct aglpair adobeglyphlist[] =
+ if (strstr(name, "u") == name)
+ return strtol(name + 1, 0, 16);
+
+ return -1;
+}
+
+struct pdf_aglpair pdf_adobeglyphlist[] =
{
{"A",0x0041},
@@ -4303,3 +4328,5 @@ static struct aglpair adobeglyphlist[] =
*/
};
+int pdf_adobeglyphlen = (sizeof(pdf_adobeglyphlist) / sizeof(struct pdf_aglpair));
+
diff --git a/mupdf/fontenc.h b/mupdf/fontenc.c
index d64e3e49..cdf082d7 100644
--- a/mupdf/fontenc.h
+++ b/mupdf/fontenc.c
@@ -2,9 +2,29 @@
* Built-in font tables
*/
+#include <fitz.h>
+#include <mupdf.h>
+
#define _notdef 0
-static char *macroman[256] = { _notdef, _notdef,
+void pdf_loadencoding(char **estrings, char *encoding)
+{
+ char **bstrings = nil;
+ int i;
+
+ if (!strcmp(encoding, "MacRomanEncoding"))
+ bstrings = pdf_macroman;
+ if (!strcmp(encoding, "MacExpertEncoding"))
+ bstrings = pdf_macexpert;
+ if (!strcmp(encoding, "WinAnsiEncoding"))
+ bstrings = pdf_winansi;
+
+ if (bstrings)
+ for (i = 0; i < 256; i++)
+ estrings[i] = bstrings[i];
+}
+
+char *pdf_macroman[256] = { _notdef, _notdef,
_notdef, _notdef, _notdef, _notdef, _notdef, _notdef,
_notdef, _notdef, _notdef, _notdef, _notdef, _notdef,
_notdef, _notdef, _notdef, _notdef, _notdef, _notdef,
@@ -44,7 +64,7 @@ static char *macroman[256] = { _notdef, _notdef,
"Ugrave", "dotlessi", "circumflex", "tilde", "macron", "breve",
"dotaccent", "ring", "cedilla", "hungarumlaut", "ogonek", "caron" };
-static char *macexpert[256] = { _notdef, _notdef,
+char *pdf_macexpert[256] = { _notdef, _notdef,
_notdef, _notdef, _notdef, _notdef, _notdef, _notdef,
_notdef, _notdef, _notdef, _notdef, _notdef, _notdef,
_notdef, _notdef, _notdef, _notdef, _notdef, _notdef,
@@ -97,7 +117,7 @@ static char *macexpert[256] = { _notdef, _notdef,
"msuperior", "commasuperior", "periodsuperior", "Dotaccentsmall",
"Ringsmall", _notdef, _notdef, _notdef, _notdef };
-static char *winansi[256] = { _notdef, _notdef, _notdef,
+char *pdf_winansi[256] = { _notdef, _notdef, _notdef,
_notdef, _notdef, _notdef, _notdef, _notdef, _notdef,
_notdef, _notdef, _notdef, _notdef, _notdef, _notdef,
_notdef, _notdef, _notdef, _notdef, _notdef, _notdef,
@@ -140,7 +160,7 @@ static char *winansi[256] = { _notdef, _notdef, _notdef,
#if 0
-static char *standard[256] = { _notdef, _notdef,
+char *pdf_standard[256] = { _notdef, _notdef,
_notdef, _notdef, _notdef, _notdef, _notdef, _notdef,
_notdef, _notdef, _notdef, _notdef, _notdef, _notdef,
_notdef, _notdef, _notdef, _notdef, _notdef, _notdef,
@@ -180,7 +200,7 @@ static char *standard[256] = { _notdef, _notdef,
_notdef, "dotlessi", _notdef, _notdef, "lslash", "oslash",
"oe", "germandbls", _notdef, _notdef, _notdef, _notdef };
-static char *expert[256] = { _notdef, _notdef, _notdef,
+char *pdf_expert[256] = { _notdef, _notdef, _notdef,
_notdef, _notdef, _notdef, _notdef, _notdef, _notdef,
_notdef, _notdef, _notdef, _notdef, _notdef, _notdef,
_notdef, _notdef, _notdef, _notdef, _notdef, _notdef,
@@ -233,7 +253,7 @@ static char *expert[256] = { _notdef, _notdef, _notdef,
"Uacutesmall", "Ucircumflexsmall", "Udieresissmall", "Yacutesmall",
"Thornsmall", "Ydieresissmall" };
-static char *symbol[256] = { _notdef, _notdef, _notdef,
+char *pdf_symbol[256] = { _notdef, _notdef, _notdef,
_notdef, _notdef, _notdef, _notdef, _notdef, _notdef,
_notdef, _notdef, _notdef, _notdef, _notdef, _notdef,
_notdef, _notdef, _notdef, _notdef, _notdef, _notdef,
@@ -280,7 +300,7 @@ static char *symbol[256] = { _notdef, _notdef, _notdef,
"bracketrightex", "bracketrightbt", "bracerighttp", "bracerightmid",
"bracerightbt", _notdef };
-static char *zapfdingbats[256] = { _notdef, _notdef,
+char *pdf_zapfdingbats[256] = { _notdef, _notdef,
_notdef, _notdef, _notdef, _notdef, _notdef, _notdef,
_notdef, _notdef, _notdef, _notdef, _notdef, _notdef,
_notdef, _notdef, _notdef, _notdef, _notdef, _notdef,
diff --git a/mupdf/interpret.c b/mupdf/interpret.c
index 5146bb2b..e377c5b6 100644
--- a/mupdf/interpret.c
+++ b/mupdf/interpret.c
@@ -82,8 +82,6 @@ runxobject(pdf_csi *csi, pdf_xref *xref, pdf_xobject *xobj)
fz_node *transform;
fz_file *file;
-puts("run xobject");
-
/* gsave */
if (csi->gtop == 31)
return fz_throw("gstate overflow in content stream");
diff --git a/mupdf/type3.c b/mupdf/type3.c
index eb403db4..60d7c38c 100644
--- a/mupdf/type3.c
+++ b/mupdf/type3.c
@@ -5,27 +5,6 @@
extern pdf_font *pdf_newfont(char *name);
-/* TODO: factor out loadencoding which is common with simple fonts */
-
-#include "fontenc.h"
-
-static void loadencoding(char **estrings, char *encoding)
-{
- char **bstrings = nil;
- int i;
-
- if (!strcmp(encoding, "MacRomanEncoding"))
- bstrings = macroman;
- if (!strcmp(encoding, "MacExpertEncoding"))
- bstrings = macexpert;
- if (!strcmp(encoding, "WinAnsiEncoding"))
- bstrings = winansi;
-
- if (bstrings)
- for (i = 0; i < 256; i++)
- estrings[i] = bstrings[i];
-}
-
static void
t3dropfont(fz_font *font)
{
@@ -124,6 +103,7 @@ pdf_loadtype3font(pdf_font **fontp, pdf_xref *xref, fz_obj *dict)
{
fz_error *error;
char buf[256];
+ unsigned short *utable;
char *estrings[256];
pdf_font *font;
fz_obj *encoding;
@@ -188,7 +168,7 @@ printf(" matrix [%g %g %g %g %g %g]\n",
goto cleanup;
if (fz_isname(obj))
- loadencoding(estrings, fz_toname(encoding));
+ pdf_loadencoding(estrings, fz_toname(encoding));
if (fz_isdict(encoding))
{
@@ -196,7 +176,7 @@ printf(" matrix [%g %g %g %g %g %g]\n",
base = fz_dictgets(encoding, "BaseEncoding");
if (fz_isname(base))
- loadencoding(estrings, fz_toname(base));
+ pdf_loadencoding(estrings, fz_toname(base));
diff = fz_dictgets(encoding, "Differences");
if (fz_isarray(diff))
@@ -223,6 +203,28 @@ printf(" matrix [%g %g %g %g %g %g]\n",
goto cleanup;
/*
+ * ToUnicode
+ */
+
+ utable = fz_malloc(sizeof(unsigned short) * 256);
+ if (!utable)
+ goto cleanup;
+
+ for (i = 0; i < 256; i++)
+ if (estrings[i])
+ utable[i] = pdf_lookupagl(estrings[i]);
+ else
+ utable[i] = i;
+
+ if (fz_dictgets(dict, "ToUnicode"))
+ {
+printf(" load tounicode cmap for type3 font\n");
+ }
+
+ font->ncidtoucs = 256;
+ font->cidtoucs = utable;
+
+ /*
* Widths
*/
diff --git a/object/simple.c b/object/simple.c
index aa736b8e..629fdf85 100644
--- a/object/simple.c
+++ b/object/simple.c
@@ -189,7 +189,7 @@ fz_toname(fz_obj *obj)
{
if (fz_isname(obj))
return obj->u.n;
- return nil;
+ return "";
}
char *
@@ -197,7 +197,7 @@ fz_tostringbuf(fz_obj *obj)
{
if (fz_isstring(obj))
return obj->u.s.buf;
- return nil;
+ return "";
}
int
diff --git a/test/pdfrip.c b/test/pdfrip.c
index 45577b82..0837e969 100644
--- a/test/pdfrip.c
+++ b/test/pdfrip.c
@@ -83,9 +83,9 @@ void dumptext(fz_node *node)
if (fabs(dy) > 1.6)
puts("\n");
- else if (fabs(dy) > 0.1)
+ else if (fabs(dy) > 0.2)
putchar('\n');
- else if (fabs(dx) > 0.1)
+ else if (fabs(dx) > 0.2)
putchar(' ');
h = fz_gethmtx(text->font, cid);
generated by cgit v1.2.3 (git 2.25.1) at 2025-03-24 14:05:41 +0000